Recombination-Based Autonomous Protein Structure-Function Search

An interesting preprint showed up last week in the autonomous agent-driven experimentation space - an area I'm paid to be very interested in.  This didn't run on our hardware, but we certainly could reproduce the experiment on our autonomous laboratory installation Nebula if the preprint had all the critical details (I'll try not to kvetch constantly about that, but it is an annoying brake on my eagerness to reproduce some or all the results).  Anyone could run it on our Cloud Lab - I haven't priced it out but this should come in at a very reasonable number.  What I found particularly clever is the solution to the problem of keeping protein engineering by AI in a reasonable cost envelope - and this is accomplished by relying on chimeras which can be a generated from one round of DNA synthesis and then clever shuffling.

Learning protein function through autonomous experimental interaction

Coban Brooks,  View ORCID ProfilePascal Notin,  View ORCID ProfilePhilip A. Romero

https://www.biorxiv.org/content/10.64898/2026.08.14.744985v1

SIMB Notebook: The Frontiers of Protein Engineering

The Society for Industrial Microbiology (SIMB) annual meeting in Austin the first week of the month was great fun.  I caught up with two Warpers, saw some other contacts from that era and met a bunch of new friends.  Plus I showed the flag for autonomous laboratories with a talk in a session on applying AI to the field - and got a chance to talk to multiple users of our systems at customer sites. Murphy's law struck in that our PR on four academic placements of autonomous labs came out about 20 hours after my talk.   In addition to the  session I appeared in, I attended a keynote on protein engineering, a session on enzyme engineering, and a whole day of talks on natural product research - plus all the posters.  I have a few notes that are maddeningly incomplete & I apologize in advance for the quality of attributions.  But here are a few observations on the space on protein engineering, which I am trying to level up to "barely dilettante" grade.

HiFi's Not Very Pacific Waters

PacBio had a major shakeup early this month. First came a large round of layoffs, which caught several individuals who were very much the face of innovation on social media.  Next came the sudden retirement of CEO Christian Henry, who has appeared in this space multiple times, and the elevation of CFO Mark Van Oene to the top seat.  

When PCR Gives You Lemons...

I wish to highlight an interesting new preprint from a group at Danish Technical University which explores the schmutz in PCR reactions, demonstrates some surprising conclusions, shows how their results might be very relevant to a common use of amplicon sequencing, and demonstrate a specific utility running PCR on in cases where typical PCR absolutely can't succeed.  Put another way, how often do people patent what was previously considered uninteresting experimental artifacts?

Dad Had Manuals, I Have An Agent

My father was a very methodical person, and that particularly applied when he decided to purchase the second IBM PC clone to appear on the market.  A whole host of manuals appeared on his desk and when I started skimming them it was clear he had been there first.  Sentences highlighted, marginal notes and of course editing marks to suit his sense of grammar and style.  The hardware, operating system, and APL programming language by far represented the most complex technology in the house - easily more complicated than everything else put together.

Element's Curious Patentsieg

Last week news broke that Element Biosciences had won a patent lawsuit against Illumina, and that it affected MiSeq.  MiSeq???? How did a patent battle in 2026 affect a platform launched nearly a decade and a half ago?  And which had been launched about five years before Element incorporated.  Which is being put out to pasture, though the MiSeqDx version What witchcraft is this?  I've done limited digging - I really dislike trying to dig through patents - and it is a strange case.

Fiber-Seq: Plasmids Model Chromosomes Well

A recent publication in Molecular Cell involving Fiber-Seq caught my attention, though I've since discovered this preprinted over a year ago.  One of the challenges in exploring the effects of genetic variations is the cost & effort to generate collections of variation which can be profiled by available methods.  This paper explores using Fiber-Seq on transfected plasmids and finds that, with important caveats, transfected plasmids evoke the same patterns of chromatin protection as the corresponding nuclear DNA.  This opens up the possibility of exploring large sets of variants for their effect on chromatin structure and either feeding that data into models or using it to understand clinical non-coding variation.  Or why not both?